Update for an Issue #2384

lib/controller/checks.py

@@ -20,6 +20,7 @@ from lib.core.common import extractRegexResult
 from lib.core.common import extractTextTagContent
 from lib.core.common import findDynamicContent
 from lib.core.common import Format
+from lib.core.common import getFilteredPageContent
 from lib.core.common import getLastRequestHTTPError
 from lib.core.common import getPublicTypeMembers
 from lib.core.common import getSafeExString
@@ -63,6 +64,7 @@ from lib.core.exception import SqlmapConnectionException
 from lib.core.exception import SqlmapNoneDataException
 from lib.core.exception import SqlmapSilentQuitException
 from lib.core.exception import SqlmapUserQuitException
+from lib.core.settings import CANDIDATE_SENTENCE_MIN_LENGTH
 from lib.core.settings import DEFAULT_GET_POST_DELIMITER
 from lib.core.settings import DUMMY_NON_SQLI_CHECK_APPENDIX
 from lib.core.settings import FI_ERROR_REGEX
@@ -478,6 +480,26 @@ def checkSqlInjection(place, parameter, value):
 
                                     injectable = True
 
+                                elif threadData.lastComparisonRatio > UPPER_RATIO_BOUND and not any((conf.string, conf.notString, conf.regexp, conf.code, kb.nullConnection)):
+                                    originalSet = set(getFilteredPageContent(kb.pageTemplate, True, "\n").split("\n"))
+                                    trueSet = set(getFilteredPageContent(truePage, True, "\n").split("\n"))
+                                    falseSet = set(getFilteredPageContent(falsePage, True, "\n").split("\n"))
+
+                                    if originalSet == trueSet != falseSet:
+                                        candidates = trueSet - falseSet
+
+                                        if candidates:
+                                            candidates = sorted(candidates, key=lambda _: len(_))
+                                            for candidate in candidates:
+                                                if re.match(r"\A[\w.,! ]+\Z", candidate) and ' ' in candidate and len(candidate) > CANDIDATE_SENTENCE_MIN_LENGTH:
+                                                    conf.string = candidate
+                                                    injectable = True
+
+                                                    infoMsg = "%s parameter '%s' appears to be '%s' injectable (with --string=\"%s\")" % (paramType, parameter, title, repr(conf.string).lstrip('u').strip("'"))
+                                                    logger.info(infoMsg)
+
+                                                    break
+
                             if injectable:
                                 if kb.pageStable and not any((conf.string, conf.notString, conf.regexp, conf.code, kb.nullConnection)):
                                     if all((falseCode, trueCode)) and falseCode != trueCode:

lib/core/common.py

@@ -1755,7 +1755,7 @@ def safeStringFormat(format_, params):
                     break
     return retVal
 
-def getFilteredPageContent(page, onlyText=True):
+def getFilteredPageContent(page, onlyText=True, split=" "):
     """
     Returns filtered page content without script, style and/or comments
     or all HTML tags
@@ -1768,10 +1768,10 @@ def getFilteredPageContent(page, onlyText=True):
 
     # only if the page's charset has been successfully identified
     if isinstance(page, unicode):
-        retVal = re.sub(r"(?si)<script.+?</script>|<!--.+?-->|<style.+?</style>%s" % (r"|<[^>]+>|\t|\n|\r" if onlyText else ""), " ", page)
-        while retVal.find("  ") != -1:
-            retVal = retVal.replace("  ", " ")
-        retVal = htmlunescape(retVal.strip())
+        retVal = re.sub(r"(?si)<script.+?</script>|<!--.+?-->|<style.+?</style>%s" % (r"|<[^>]+>|\t|\n|\r" if onlyText else ""), split, page)
+        while retVal.find(2 * split) != -1:
+            retVal = retVal.replace(2 * split, split)
+        retVal = htmlunescape(retVal.strip().strip(split))
 
     return retVal
 

lib/core/settings.py

@@ -19,7 +19,7 @@ from lib.core.enums import DBMS_DIRECTORY_NAME
 from lib.core.enums import OS
 
 # sqlmap version (<major>.<minor>.<month>.<monthly commit>)
-VERSION = "1.1.2.3"
+VERSION = "1.1.2.4"
 TYPE = "dev" if VERSION.count('.') > 2 and VERSION.split('.')[-1] != '0' else "stable"
 TYPE_COLORS = {"dev": 33, "stable": 90, "pip": 34}
 VERSION_STRING = "sqlmap/%s#%s" % ('.'.join(VERSION.split('.')[:-1]) if VERSION.count('.') > 2 and VERSION.split('.')[-1] == '0' else VERSION, TYPE)
@@ -359,6 +359,9 @@ MIN_RATIO = 0.0
 # Maximum value for comparison ratio
 MAX_RATIO = 1.0
 
+# Minimum length of sentence for automatic choosing of --string (in case of high matching ratio)
+CANDIDATE_SENTENCE_MIN_LENGTH = 10
+
 # Character used for marking injectable position inside provided data
 CUSTOM_INJECTION_MARK_CHAR = '*'
 

lib/core/threads.py

@@ -46,6 +46,7 @@ class _ThreadData(threading.local):
         self.lastComparisonPage = None
         self.lastComparisonHeaders = None
         self.lastComparisonCode = None
+        self.lastComparisonRatio = None
         self.lastErrorPage = None
         self.lastHTTPError = None
         self.lastRedirectMsg = None

lib/request/comparison.py

@@ -144,6 +144,9 @@ def _comparison(page, headers, code, getRatioValue, pageLength):
             kb.matchRatio = ratio
             logger.debug("setting match ratio for current parameter to %.3f" % kb.matchRatio)
 
+    if kb.testMode:
+        threadData.lastComparisonRatio = ratio
+
     # If it has been requested to return the ratio and not a comparison
     # response
     if getRatioValue: